Simulation Hypothesis

[UnReAL13]

On the quantum level, the particles of reality differ in no way from each other. Whether it's a hydrogen atom or a helium atom, the basic quantum units are fundamentally the same. The composition of the atoms only depends on the amount of each kind particle each one contains. Almost like an infinitely complex system of "coding".

This has lead some people to the ultimate conclusion that our known reality is actually the most realistic simulation ever created. It would be indistinguishable from the real "reality" on almost every observable level. But the quantum level is where the strangest phenomena occurs, and is vastly inconsistent with standard physics.

There is certainly other quantum phenomena that seems to indicate this hypothesis. The fact that every particle collapses from a wave to a particle only when observed, the observation or measurement problem. This is easily consistent with the concept of algorithms breaking down into single pixels. There's also quantum entanglement, the replication of one particle's activity in another one no matter how far apart the distance. This is easily consistent with the concept of pixels of data placed across the screen from each other, yet their programming is stored in the memory where space and time have no meaning. This is all explained much more in depth in this person's essay: Quantum Mechanics Implies That The Universe is a Computer Simulation

If Moore's Law continues, then we will eventually end up building a computer with an infinite amount of processing power. This will be able to produce an infinite amount of bits of data, thus enabling the most realistic simulations imaginable. It's quite possible that at some point in the future people will want to run ancestor simulations to discover the origins of their bloodlines. Some people say that based on the laws of probability, if an infinite number of simulations could be created, then the chances that we're living in the true "reality" are slim to none. It would be almost inevitable for much of this to occur, according to some.

The strangeness of quantum reality has only raised so many more questions. Could this be the one hypothesis that is consistent with all quantum mechanics? Perhaps further venture into this field of science could reveal the most inconvenient truth of them all



Just remember, this hypothesis is based on the assumption of 3 possibilities:

1) The human race will go extinct before reaching a post-human stage.

2) The human race will reach a post-human stage, but won't be able to run a significant number of ancestor simulations.

3) We are most certainly living inside of an ancestor simulation right now.


If the human race was able to run a significant number of simulations, perhaps an infinite amount, then the probability of the 3rd possibility would be astronomically higher than the first two. All worth considering I suppose, and I might even go as far as labeling this a "theory" depending on further investigation of quantum phenomena


Here's a good short video synopsis as well:

YouTube - Do we really exist?

 

[AlexBP]

I actually came up with this argument while I was in college and was very impressed with my argument for a while. Nowadays I'm a Christian so I don't believe this sort of reasoning. However, if I were strictly within the realm of materialist thought I could still find a flaw in the reasoning.

Suppose there were an infinite number of worlds that were real. Suppose that within that infinite number of worlds, an infinite number had beings intelligent enough to create simulations. Suppose that each world in which simulations were created, the number of simulations were large, or perhaps even infinite. That would mean that in the entire cosmos there were more simulations than realities, right?

Actually not, because infinity multiplied by anything is still infinity, no more, no less. In fact, infinity multiplied by infinity is infinity. In mathematical terms, we would say that the cardinality of the number of realities and the number of simulations is the same. Therefore we aren't actually more likely to be living in a simulation at all.

 

[Bushido216]

Your counter argument breaks down in a circumstance where the numbers are finite.

 

[Gracchus]

I actually came up with this argument while I was in college and was very impressed with my argument for a while. Nowadays I'm a Christian so I don't believe this sort of reasoning. However, if I were strictly within the realm of materialist thought I could still find a flaw in the reasoning.

Suppose there were an infinite number of worlds that were real. Suppose that within that infinite number of worlds, an infinite number had beings intelligent enough to create simulations. Suppose that each world in which simulations were created, the number of simulations were large, or perhaps even infinite. That would mean that in the entire cosmos there were more simulations than realities, right?

Actually not, because infinity multiplied by anything is still infinity, no more, no less. In fact, infinity multiplied by infinity is infinity. In mathematical terms, we would say that the cardinality of the number of realities and the number of simulations is the same. Therefore we aren't actually more likely to be living in a simulation at all.

You do know don't you that some infinities are greater than others. The mathematician Georg Cantor proved that the number of real numbers is greater thant the number of integers. In fact there are an infinite number of infinities, which may be ordered from lesser to greater.
That is counter-intuitive but the mathematical proof is there for those who care to plumb the depths of set theory.

 

[GrowingSmaller]

An infinite number? IFAIK infinity is not a point on a number line.

Also we could be in a simulation, but I think that it potentially complicates the model of the universe without adding explanatory power. But why should a theist mention Occam's Razor anyway?

As for the points about pixels and algorithms, I am not eductated enough to try for an opinion.

I recall reading about a book a few years ago though, via a educational division of Yahoo I cant seem to find, and the author's theory was that certain aspects of reality were best described as programmes.

 

[AlexBP]

You do know don't you that some infinities are greater than others. The mathematician Georg Cantor proved that the number of real numbers is greater thant the number of integers. In fact there are an infinite number of infinities, which may be ordered from lesser to greater.
That is counter-intuitive but the mathematical proof is there for those who care to plumb the depths of set theory.

If the number of of real worlds is countably infinite, then the number of simulations will be countably infinite. If the number of real worlds is uncountable, then the number of simulations will be uncountable. Whatever cardinality the set of real worlds has, the set of simulations will have that same cardinality. The only exception being with finite numbers, as bushido pointed out.

 

[Chatter]

If the number of of real worlds is countably infinite, then the number of simulations will be countably infinite. If the number of real worlds is uncountable, then the number of simulations will be uncountable. Whatever cardinality the set of real worlds has, the set of simulations will have that same cardinality. The only exception being with finite numbers, as bushido pointed out.

Right. If we have a set of worlds with cardinality n, and on each of these worlds, we run a set of simulations with cardinality m, then we have n x m simulations. When cardinalities become infinite, taking a product like this is exactly the same as taking the maximum. That means you can't get any larger infinity out of the process.

Cantor's example of there being more real numbers than natural numbers is not due to taking a product, but an exponentiation. Specifically, if you encode every real number in binary form, then you are computing 2[sup]N[/sup], which is strictly greater than N.

In terms of probability, and the question "are we likely to be in a simulation", when you start having an infinite sample space, you need to name your probability distribution. There's no "fair" distribution in an arbitrary infinite space.

My pessimism about the future of technology and simulation in particular says that we won't be building any simulations at all.

 

[GrowingSmaller]

There are certain simulations we have already, like the 'millenium run' in astrophysics, and Alife in biology.

 

[Chatter]

There are certain simulations we have already, like the 'millenium run' in astrophysics, and Alife in biology.

There are certain simulations in the computer game Half-Life 2. They're just as irrelevant to the simulation argument.

 

[GrowingSmaller]

What I mean is that physical systems can be modelled on a computer. If the brain, and it's experience is physical, then in principle it can be modelled if a) we know enough about how it works, and b) we have the computing skills. Isn't that what artificial intelligence, and consequently simulation, is at least partially about?

My pessimism about the future of technology and simulation in particular says that we won't be building any simulations at all.

Not even the most basic? They might already exist...

18 November 2009—Scientists and engineers at IBM’s Almaden Research Center, in San Jose, Calif., announced today at the Supercomputing Conference (SC09) in Portland, Ore., that they have created the largest brain simulation to date on a supercomputer.

 

[Chatter]

What I mean is that physical systems can be modelled on a computer. If the brain, and it's experience is physical, then in principle it can be modelled if a) we know enough about how it works, and b) we have the computing skills. Isn't that what artificial intelligence, and consequently simulation, is at least partially about?

Not really. Artificial intelligence is predominantly concerned with computational symbolic models of brain processes, rather than simulations. The people doing simulations tend to be from neuroscience backgrounds. See the Blue-Brain Project.

Not even the most basic? They might already exist...

Again, not in the relevant sense. We're not talking about merely doing brain simulation, something which goes on just down the corridor from my office. We're talking about being able to simulate ourselves on a massive scale. That's what I'm sceptical about. Moore's Law predicts exponential growth in computing power, but it has limits, and those might fall far short of what is required for the Simulation Hypothesis. That's my feeling.

 

[GrowingSmaller]

If Moore's Law continues, then we will eventually end up building a computer with an infinite amount of processing power. This will be able to produce an infinite amount of bits of data, thus enabling the most realistic simulations imaginable.

I thought that Moores Law fits a nomal curve of finite proportins. So, at time t, whatever that time (100,10000, or 1000,000 years time etc) the processing power achieved will always be finite.

As for having an infinite amount of bits of data, I think that in that case the computer would have to be equinumerous in it's computations to the 'computations' of the whole universe (assuming that both the parts and the whole of the universe are infinite), and we would have to have stable computations from every part of the computer subsystem, down to the superstring and infinitely beyond. Now that would really be a "supercomputer"! Or something along those lines, and I don't think that is achievable.

 

[AlexBP]

Right. If we have a set of worlds with cardinality n, and on each of these worlds, we run a set of simulations with cardinality m, then we have n x m simulations. When cardinalities become infinite, taking a product like this is exactly the same as taking the maximum. That means you can't get any larger infinity out of the process.

Cantor's example of there being more real numbers than natural numbers is not due to taking a product, but an exponentiation. Specifically, if you encode every real number in binary form, then you are computing 2[sup]N[/sup], which is strictly greater than N.

In terms of probability, and the question "are we likely to be in a simulation", when you start having an infinite sample space, you need to name your probability distribution. There's no "fair" distribution in an arbitrary infinite space.

That's exactly what I was trying to say, that you phrased it much better than I did. However, the mathematical rebuttal to the original post in this thread is something of a diversion, although an amusing one. In truth, we know we're not in a simulation because we have free will, whereas a simulation would not.

However, if you want a really twisty idea, consider this. Some people take seriously the notion that the "reality" we humans perceive is a simulation akin to tht generated by computers. We can contemplate this idea know, but no one could contemplate it prior to the invention of computers. So perhaps the true, metaphysical notion of reality is actually of some form that we can't even contemplate yet, because the appropriate technology hasn't been invented yet.

 

[Bushido216]

That's exactly what I was trying to say, that you phrased it much better than I did. However, the mathematical rebuttal to the original post in this thread is something of a diversion, although an amusing one. In truth, we know we're not in a simulation because we have free will, whereas a simulation would not.

However, if you want a really twisty idea, consider this. Some people take seriously the notion that the "reality" we humans perceive is a simulation akin to tht generated by computers. We can contemplate this idea know, but no one could contemplate it prior to the invention of computers. So perhaps the true, metaphysical notion of reality is actually of some form that we can't even contemplate yet, because the appropriate technology hasn't been invented yet.

At the risk of sounding nitpicky, provided that the times we're talking about aren't so sufficiently large that there could potentially be a big shift in brain capacity thanks to evolution, anything that we could think about in the future we can think about now. Adding computers didn't add to the list of things we can conceptualize about, at best it gave us tools and the paradigm to think about simulations.

 

[GrowingSmaller]

We're talking about being able to simulate ourselves on a massive scale. That's what I'm sceptical about.

I am skeptical of that too. But i reckion that some form of simulation of experience might be possible, or even already be happening (maybe just down from your office) without us knowing it.

 

[GrowingSmaller]

There is certainly other quantum phenomena that seems to indicate this hypothesis. The fact that every particle collapses from a wave to a particle only when observed, the observation or measurement problem. This is easily consistent with the concept of algorithms breaking down into single pixels.

There's also quantum entanglement, the replication of one particle's activity in another one no matter how far apart the distance. This is easily consistent with the concept of pixels of data placed across the screen from each other, yet their programming is stored in the memory where space and time have no meaning.

This is all explained much more in depth in this person's essay: Quantum Mechanics Implies That The Universe is a Computer Simulation

I will have to read that soon. Anyone any thoughts on it so far?

 

[sandwiches]

If Moore's Law continues, then we will eventually end up building a computer with an infinite amount of processing power.

This reminds me of the Omega Particle nonsense. And this is wrong. Moore's Law in no way indicates that we can ever build a computer with infinite amount of processing power. It's like saying that we'll eventually be able to count all the 9s in the number .99999... simply because we're able to count exponentially faster and faster.

 

[AlexBP]

Good metaphor.

 

[Heavens]

I exist!
Oh, wait, sorry... not sure now, all the data changed when I looked to close